Processor and processor method of operation

Abstract

In one embodiment, the present invention is directed to a processor that comprises an instruction pipeline for executing processor instructions wherein the processor instructions define a memory access size and a cache memory for storing cache lines in a plurality of memory banks that have a block size that is greater than the memory access size, the cache memory including mapping logic for storing contiguous groups of bits, of size equal to the memory access size, in different ones of the plurality of memory banks.

1. A processor, comprising:
 an instruction pipeline for executing processor instructions wherein said processor instructions define a memory access size; and 
 a cache memory for storing cache lines in a plurality of memory banks that have a block size that is greater than said memory access size, said cache memory including mapping logic for storing contiguous groups of bits, of size equal to said memory access size, in different ones of said plurality of memory banks. 
 
   
   
     2. The processor of  claim 1 , wherein said instruction pipeline comprises:
 a plurality of functional units for executing said processor instructions. 
 
   
   
     3. The processor of  claim 2  wherein said plurality of functional units are operable to initiate a memory access to a first memory address of data stored in said cache memory during pipeline execution of a first processor instruction while memory access to a second memory address that is immediately before said first memory address is occurring during pipeline execution of a second processor instruction. 
   
   
     4. The processor of  claim 3  wherein said first processor instruction is obtained by an instruction fetch unit while said second processor instruction is being processed in said instruction pipeline. 
   
   
     5. The processor of  claim 1  wherein said cache memory obtains cache lines from main memory. 
   
   
     6. The processor of  claim 1  wherein said cache memory obtains cache lines from a lower level cache. 
   
   
     7. The processor of  claim 1  wherein said mapping logic stores adjacent words in said cache lines in different ones of said plurality of memory banks. 
   
   
     8. A method for operating a processor, comprising:
 executing instructions in an instruction pipeline, wherein said instructions define a memory access size; 
 storing cache lines in memory banks of cache memory included in said processor, said memory banks having a block size that is greater than said memory access size, wherein said storing causes contiguous groups of bits, of size equal to said memory access size, to be stored in different ones of said memory banks. 
 
   
   
     9. The method of  claim 8  wherein said executing instruction in said instruction pipeline includes:
 executing instructions in a plurality of functional units in parallel. 
 
   
   
     10. The method of  claim 8  further including:
 initiating a first memory access to a first memory address of data stored in said cache memory during pipeline execution of a first instruction; and 
 initiating a second memory access to a second memory address of data stored in said cache memory during pipeline execution of a second instruction while said first memory access is pending, wherein said second memory address is immediately after said first memory address by a number of bits equal to said memory access size. 
 
   
   
     11. The method of  claim 10  further comprising:
 fetching said second instruction while said first memory access is pending. 
 
   
   
     12. The method of  claim 8  further comprising:
 retrieving cache lines from main memory; and 
 storing said cache lines in said cache memory. 
 
   
   
     13. The method of  claim 8  further comprising:
 retrieving cache lines from a lower level cache; and 
 storing said cache lines in said cache memory. 
 
   
   
     14. The method of  claim 8  wherein said storing stores adjacent words in said cache lines in different ones of said memory banks. 
   
   
     15. A processor, comprising:
 a pipeline means for executing processor instructions wherein said processor instructions define a memory access size; and 
 a cache memory means for storing cache lines in memory banks according to a block size that is greater than said memory access size, wherein said cache memory means stores contiguous groups of bits, of size equal to said memory access size, in different ones of said plurality of memory banks. 
 
   
   
     16. The processor of  claim 15  wherein said pipeline means includes:
 a plurality of functional units for executing said processor instructions in parallel. 
 
   
   
     17. The processor of  claim 16  wherein said plurality of functional units are operable to initiate a memory access to a first memory address of data stored in said cache memory means during pipeline execution of a first processor instruction while memory access to a second memory address that is immediately before said first memory address is occurring during pipeline execution of a second processor instruction. 
   
   
     18. The processor of  claim 17  further comprising:
 an instruction fetch means for fetching said first and second processor instructions, wherein said instruction fetch means fetches said first processor instruction while said second memory access is occurnng. 
 
   
   
     19. The processor of  claim 15  wherein said cache memory means retrieves said cache lines from main memory. 
   
   
     20. The processor of  claim 15  wherein said cache memory means stores adjacent words of said cache lines in different ones of said memory banks.